Ludlow (Silurian) stromatoporoid biostromes from Gotland, Sweden: facies, depositional models and modern analogues

Stacked stromatoporoid‐dominated biostromes of the Ludlow‐age Hemse Group (Silurian) in eastern Gotland, Sweden, are 0·5–5 m thick and a few tens of metres to >1 km in lateral extent. They form one of the world's richest Palaeozoic stromatoporoid deposits. This study compiles published and new data to provide an overall facies model for these biostromes, which is assessed in relation to possible modern analogues. Some biostromes have predominantly in‐place fossils and are regarded as reefs, but lack rigid frameworks because of abundant low‐profile non‐framebuilding stromatoporoids; other biostromes consist of stromatoporoid‐rich rudstones interpreted here as storm deposits. Variation between these two `end‐members' occurs both between interlayered biostromes and also vertically and laterally within individual biostromes. Such variation produces problems of applying established reef classification terms and demonstrates the need for the development of terminology that recognizes taphonomic destruction of reef fabrics. An approach to such terminology is found in all four categories of a recent biostrome classification scheme that are easily recognized in the Hemse biostrome facies: autobiostromes (>60% in place); autoparabiostromes (a mixture of in‐place and overturned reef‐building organisms, 20–60% in place); parabiostromes (builders are overturned and damaged, <20% in place); and allobiostromes (transported and detrital reef material, nothing in place). These categories provide a broad taphofacies scheme for the Hemse biostromes, which are mostly autoparabiostrome to allobiostrome. The biostromes developed on crinoidal grainstone sheets and expanded laterally across relatively flat substrates in a marine setting of low siliciclastic input. Planar erosion surfaces commonly terminate biostrome tops. Three broadly similar modern analogues are identified, each of which has elements in common with the Hemse biostromes, but none of which is an exact equivalent: (a) laterally expanded and coalesced back‐barrier patch reefs behind the Belize barrier, an area influenced by limited accommodation space; (b) a hurricane‐influenced shelf, interpreted for Grand Cayman, where reef cores consist of rubble and lack substantial framework; the wide distribution of rounded pebbles and cobbles of stromatoporoids in the Hemse biostromes most probably resulted from hurricanes; (c) coral carpets in 5–15 m water depth of the northern Red Sea, where lateral expansion of low‐diversity frames dominated by Porites coral has produced low‐profile biostromes up to 8 m thick and several km long. Such carpets accumulated large amounts of carbonate, with little export, as in the Hemse biostromes, although the latter did not build frameworks because of the nature of growth of the stromatoporoids. The notable lack of algae in the Hemse biostrome facies is also a feature of Red Sea coral carpets; nevertheless, coral carpets are ecologically different. Hemse biostromes lack evidence of a barrier reef system, although this may not be exposed; the facies assemblage is consistent with either a storm/hurricane‐influenced mid‐ to upper ramp or back‐barrier system.